Back to Home   Several VoIP Audio
Questions and Answers
By Kevin Custer W3KKC 		  Print this Page


Pre-Emphasis and De-Emphasis: Done in the radio or done by Allstar?

Pre-emphasis and de-emphasis have to be done once only... somewhere. I prefer to let AllStar do the work, because all nodes will sound alike even if the repeater equipment is different, as long as the frequency response of the repeater radios extends beyond that of the AllStar audio path. Audio levels are adjusted by the "Radio Tune" menu and can be controlled to give the perfect amount depending on what the repeater transmitter requires.

The options for transmitter audio are:

  1. Normal audio (no pre-emphasis or limiting).
  2. Pre-emphasized audio without limiting.
  3. Pre-emphasized audio with limiting.

Option 1 is used with MIC audio inputs where the radio will perform the pre-emphasis and limiting. Option 2 is used where the transmitter requires pre-emphasized audio but its audio circuit provides limiting. Option 3 is used where you are driving the FM modulator directly and no processing of any kind exists in the transmitter audio chain.

There are two transmitter audio outputs from most radio adapters (Main & Aux or Left & Right) that can be programmed to output several combinations of transmitter audio. Either output can consist of voice, PL tone, neither, or a combination of voice and PL. You wouldn't use combined audio to feed a MIC input because the limiter would strip the PL tone in loud passages. Voice audio can be put into a MIC input or directly into a modulator. If the repeater transmitter has dual modulators (many do), I use one output set to feed voice audio to one modulator, and the other output to feed the PL tone (encode) signaling to the other modulator. If the transmitter has only one FM modulator, then a combination of voice audio and PL tone can be fed in. Of course, pre-emphasis and limiting must be applied in the "config" file. This would only affect the voice part of the channel driver.

For receiver audio, you can get audio from the speaker or discriminator/detector. When using speaker audio, the radio has already de-emphasized the audio and filtered it beyond 3000 Hz. As such, "simpleUSB" doesn't add de-emphasis by default, and you should only use speaker audio with "simpleUSB" because there isn't enough noise energy for proper noise squelch determination in "USBradio". Also, the speaker audio path in many radios filters out the PL tone so you cannot decode a PL tone using this audio source.

Conversely, when using discriminator audio and "USBradio" (what I call full DSP) the application can figure out the noise squelch and PL tone detection. This is the mode I always use. By default, "USBradio" applies other audio filtering, like high pass (300 Hz) filtering to eliminate PL pass-through to the local or remote transmitters, limiting (if selected), and low pass (3400 Hz) splatter filtering. Since all filtering and processing is identical, nodes will sound alike even if the repeater radio equipment is different.

I know this is a lot to digest, but since the "app_rpt" application (module) running under Asterisk is so versatile, a complete explanation is necessary.



Telemetry Ducking: What is it?

First, let me define what Telemetry is, in the AllStar Link system. Telemetry is anything that the node box generates to indicate something. This includes, but is not limited to, CW ID tones, courtesy tones, and voice responses from Allison or any other voice generator.

Telemetry Ducking is a features set that does two things:

  1. The setting of "telemnom=X" in some software packages allows the node operator to set their desired level of telemetry independently of the voice audio through the node. It allows the normal or "nominal" volume level of these things. In my opinion, when a node is set up properly with correct audio throughput levels, telemetry, namely Allison (the female voice behind Asterisk), is too loud. Normally a negative value (in dB) is inputted. Personally I use -9.
  2. The setting of "telemduck=X" further reduces the level of the telemetry if someone - somewhere - has the PTT depressed. This could be a user locally or one on another repeater that's currently linked. Since nodes "speak their peace" at differing times, a user waiting for the ID to finish may be keying up and speaking just when the node you are listening to has a connect or disconnect message or ID playing. This feature ducks the telemetry down even farther to reduce hearing two voices at the same time, possibly making it difficult to understand the operator on the other end. Personally, I use a value of -15 here. So, in essence, if someone has the hammer down, the node going off at the lip doesn't distract from the intended communication.

Not all versions of AllStar Link have this feature set. XIPAR by Xelatec had it first, and then the HamVoip Group ported it to their distribution. ACID and DIAL don't currently have it.



Optional Audio Filters?

This PDF shows how the optional audio filter rules can improve the audio response and bandwidth of app_rpt.

Back to the top of the page
Back to Home

This page initially created 04-August-2017 by Kevin Custer W3KKC.

This web site, the information presented in and on its pages and in these modifications and conversions is © Copyrighted 1995 and (date of last update) by Kevin Custer W3KKC and multiple originating authors. All Rights Reserved, including that of paper and web publication elsewhere.